I am a Computer Science Ph.D. student at UCLA, advised by Kai-Wei Chang and Nanyun Peng. I work on Multimodal (Vision + Language) and Embodied Learning.
At Meta, I contributed to grounding post-training for Muse Models 🥑 🍉. I also built training data pipelines for Segment Anything Models and created SAM 3 Agent.
Previously, I worked on embodied agents at Stanford SVL, with Manling Li, Jiajun Wu, and Fei-Fei Li.